Sentiment Analysis of Russian Fictional Literature

Sentiment analysis is one of the fastest-growing areas of computational linguistics. This type of analysis is used to extract and study subjective information. Sentiment Analysis measures people's opinions and emotions through natural language processing and text analysis. The research is devoted to the study of the tonality of the Corpus of Russian stories of the period between 1900-1930. The corpus consists of 557 randomly selected short stories which were written during this period. This study suggests that historical events (such as war or revolution) influence the vocabulary used by authors to write stories. With the usage of the dictionary-based algorithm for automated sentiment analysis, the following numbers were determined: the overall number of words for each period, the number of emotionally charged words, and the ratio of emotionally charged words to the total number of words. The number of positive and negative words is counted and compared. As a result, the overall tonality of each period is determined.